Substrate regulation of product distribution in the reactions of

1 Nov 1991 - Charles P. Casey, Thomas E. Vos, John T. Brady, and Randy K. Hayashi. Organometallics ... Bernard L. Flynn, Heiko Schirmer, Michael Duets...
0 downloads 0 Views 3MB Size
Download PDF
J . Am. Chem. SOC.1991, 113, 9293-9319

9293

Substrate Regulation of Product Distribution in the Reactions of Aryl Chromium Carbene Complexes with Alkynes Mary Ellen Bos, William D. Wulff,* Ross A. Miller, Steven Chamberlin, and Timothy A. Brandvold Contribution from the Searle Chemistry Laboratory, Department of Chemistry, University of Chicago, Chicago, Illinois 60637. Received February 8, 1991

Abstract: The reactions of arylcarbene complexes with alkynes were examined for six of the nine possible substitution patterns for m o m and dioxygenated aryl substituents of the carbene carbon. The product distributions were found to be highly dependent on a number of factors, including solvent, temperature, concentration of alkyne, and the nature of the aryl substituent. The product distributions were determined in nearly all cases for phenol and indene products and in some cases for furans, cyclobutenones, and cyclopentenediones, which were minor products in these reactions. The product distribution for the reaction of each arylcarbene complex was determined as a function of both temperature and alkyne concentration, since the combined product distribution profiles provided a much more sensitive measure of the relative influences of the aryl substituents on the reaction outcome. Furthermore, this distribution profile was determined for the reactions with 3-hexyne and 1-pentyne for each carbene complex. A series of monosubstituted arylcarbene complexes were examined to identify the effects of oxygen substituents at various positions on the aryl ring. The m-methoxy group has no effect on the product distribution, whereas the o-methoxy group influences the distribution by its ability to chelate to the metal center and the p-methoxy group influences the distribution by its ability to donate electrons by resonance. The product distributions from the reactions of the 2,3-, 2,4-, and 2,S-dimethoxy complexes followed the profile expected from the simple sum of the profiles of the monomethoxyl complexes. In all cases where an effect was observed, higher concentrations of alkyne led to a higher selectivity for phenol over indene products. The dependence of the product distribution on the concentration of the alkyne substrate is suggested to be due to a process in which a second molecule of alkyne coordinates to the metal center and determines the chemical outcome of an intermediate that has covalently incorporated the first alkyne. It is further suggested that the special ability of an alkyne to display this effect is related to the ability of an alkyne to readily switch from a 2 to a 4 e- donor. This phenomenon of substrate regulation of product distribution is termed the allochemical effect, and a mechanistic explanation is developed that features this proposed process and that is refined to accommodate the observed effects of solvent, temperature, chelation, and steric and electronic effects that have been observed for the reaction of carbene complexes and alkynes.

The reaction of chromium Fischer carbene complexes with alkynes was first reported 15 years ago,I and since that time much has been learned about this reaction.* The reaction has been utilized in synthetic applications more than any other reaction of transition metal carbene complexes, and the full synthetic scope of this reaction has yet to be e~tablished.~All of the applications that have appeared to date have utilized the ability of this reaction to construct highly functionalized aromatic rings in regioselective fashion under neutral conditions near ambient temperature. For example, a synthesis of 1 I-deoxydaunomycinone has been reported from our laboratories where the key step is the reaction of the o-methoxy complex 2b with the alkyne 3, which occurs with high regio- and chemoselectivity in 80% yielde4q5 Presently we have several syntheses of other natural products underway in which more highly oxygenated arylcarbene complexes will be required in the key benzannulation step, including the synthesis of 7( I ) DBtz, K. H. Angew. Chem., Inr. Ed. Engl. 1975, 14, 644. (2) For reviews on the synthetic applications of Fischer carbene complexes, see: (a) Brown, E. J. frog. Inorg. Chem. 1980, 27, 1. (b) DBtz, K. H.; Fischer, H.; Hofmann, P.; Kreissel, F. R.; Schubert, U.; Weiss, K. Transition Metal Carbene Complexes; Verlag Chemie: Deerfield Beach, FL, 1984. (c) DBtz, K. H. Angew. Chem., Inr. Ed. Engl. 1984, 23, 587. (d) Casey, C. P. React. Intermed. 1985, 3. (e) Wulff, W. D.; Tang, P. C.; Chan, K. S.; McCallum, J. S.;Yang, D. C.; Gilt", S.R. Tetrahedron 1985,41,5813. (f) Chan, K. S.; Peterson, G. A,; Brandvold, T. A.; Faron, K. L.; Challener, C. A.; Hyldahl, C.; Wulff, W. D. J. Organomet. Chem. 1987, 334, 9. (g) Wtz, K. H. In Organometallics in Organic Synthesis: Aspects of a Modern Interdisciplinary Field; Dieck, H. tom, de Meijere, A., Eds.; Springer: Berlin, 1988. (h) Schore, N. E. Chem. Reo. 1988,88, 1081. (i) Advances in Metal Carbene Chemistry; Schubert, U . , Ed.; Kluwer Academic Publishers: Boston, 1989. (j)Wulff, W. D. In Aduances in Metal-Organic Chemistry; Liebeskind, L.S., Ed.; JAI Press Inc.: Greenwich, CT, 1989; Vol. I . (k) Wulff, W. D. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, l., Eds.; Per-

gamon Press: New York, 1991; Vol. 5 . (3) For a list of citations to syntheses, see: Wulff, W. D.; Bauta, W. E.; Kaesler, R. W.; Lankford, P. J.; Miller, R. A,; Murray, C. K.; Yang, D. C. J. Am. Chem. SOC.1990. 112, 3642, and ref 2k. (4) Wulff, W. D.; Xu, Y. C. J. Am. Chem. SOC.1988. 110, 2312. ( 5 ) For references to other anthracycline syntheses with carbene complexes, see the citations listed in reference 4 and in DBtz, K. H.; Popall, M. Chem. Ber. 1988, 121, 665.

0002-7863/91/1513-9293$02.50/0

con-0-methylnogarol 46 and olivin 7,' the aglycone of olivomycin A.9 Benzannulations of carbene complexes that bear more than one oxygen substituent on the aryl ring such as 5 and 9 have not been extensively studied,1° and a t the outset there was a certain degree of uncertainty with regard to the success of these proposed benzannulations (Scheme I). Introduction The uncertainty associated with the anticipated annulations of the highly substituted complexes 5 and 9 has to do with the fact that these reactions may not give rise to the exclusive formation of six-membered ring annulated phenol products. Under defined conditions, the reaction of arylcarbene complexes and alkynes is, in general, remarkably chemoselective for the formation of phenols. However, depending on the substrates and the reaction conditions, up to twelve structurally different organic products can be produced from this reaction, including 4-alkoxyphenols, indenes, furans, cyclobutenones, vinylcyclopentenediones,phenols, pyrones, vinyl ketenes, and lactones.*,"J2 The effect of introducing a methoxyl group into the ortho position of an aryl substituent of a carbene complex was known prior to this work and is illustrated in Scheme 11. The reaction of the parent phenylcarbene complex 2a with diethylacetylene gives only the quinone 14a, whereas, the more electron-rich o-methoxy complex 2b gives both quinone and (6) For initial studies, see refs 2e, 2j, and 8. (7) For initial studies, see: Peterson, G. A,; Kunng, F. A.; McCallum, J. S . ; Wulff, W. D. Tetrahedron Lett. 1987, 28, 1381, and ref 2f. (8) (a) Semmelhack, M. F.; Jeong, N. Tetrahedron Lett. 1990, 31, 605. (b) Semmelhack, M. F.; Jeong, N.; Lee, R. L. Tetrahedron Lett. 1990. 31, 609. (9) Antineoplastic Agents; Remers, W. A,, Ed.; Wiley Interscience: New York, 1984. ( I O ) The only previous examples of annulations of arylcarbene complexes with at least two oxygen substituents can be found in ref 8 and in Boger, D. L.; Jacobson, 1. C. J . Org. Chem. 1991, 56, 2115. ( 1 I ) Xu, Y. C.; Challener, C. A,; Dragisich, V.; Brandvold,T. A.; Peterson, G. A,; Wulff, W. D.; Williard, P. G. J. Am. Chem. Soc. 1989, 1 1 1 , 1269. (12) Brandvold, T. A,; Wulff, W. D.; Rheingold, A. L. J. Am. Chem. Soc. 1990, 112, 1645.

0 1991 American Chemical Society

Bos et al.

9294 J . Am. Chem. SOC.,Vol. 113, No. 24, 1991 Scheme I 0

MeO

1

0

*

0 OH OH 11deoxydaunomyanone

Cr(CO),

Me0

01-Bu

3

2b

OH

Me2N H

O

A I ' B I cT

0

MeO

4

*

w'"OH3 H

OH

OM

&I

R30

7con-O-methylnogarol

0

Cr(CO),

01-Bu

5

*

6 +

O H ,,,? ,

OM

pA9 OH

0

0

R,

R30

7 8

+ 7::

R1 = H,R2 = H o l i n R, = Me, & = H chrmomydnone

Cr(CO)5

01-Bu

10

9

Scheme 11 U

RLCtCRs

1)

(c0)5cr$i MF, (2e q450 wC

RL +

2) CB(NH4)dNq)z

R

R

Comolex 2a 2b 2a 2b

W R s +

RS

R

O

E B L B s

OM

nPr H 11. 7 3 % 120 N. D. 12b 51 % l l b 64% nPr H H Et Et 14a 8 8 % 15. 50.3 % 14b 61 V' 15b 18% Et OMe Et " This r e a c h also Qives a 11 %yieldof h e keb ester 4Ob f l a k 5). ** Not Determined (refer" 2f)

H OMe

13a N.D. ** 13b I t % 16a 50.3 % 16b 5 %

OH PL

A

b e

17

R

6Me 18

It should be noted that in Drevious indene studies2P it has been most convenient to employ an dxidative workupI4 for these reactions, in which case the phenol products 17 are typically isolated as the quinones, and the indene products 20 and 21 are isolated in either the unoxidized, oxidized (indenone), or hydrolyzed (indanone) form. The indene products differ from the phenol products only in that a CO ligand from the metal is incorporated into the phenols but not into the indenes (18 versus 19, Scheme 11). The distribution between phenol and indene products from the reaction of Zb is dependent on the nature of ( I 3) A preference for indene products has also been observed for the more electron-rich amino carbene complexes. See ref 2f and (a) Dbtz, K. H.; Pruskil, 1. Chem. Eer. 1978,111,2059. (b) Yamashita, A. Tetrohedron Lett. 1986, 27, 5915. (c) Yamashita, A,; Toy, A.; Watt, W.; Muchmore, C. R.

Tetrahedron Lett. 1988, 29, 3403.

(14) The advantages of employing an oxidative workup are that many unstable metal complexes are destroyed, liberating their organic ligands and thus simplifying separation, and that the phenol products are in some cases somewhat unstable to air and are best isolated as their quinones. The disadvantage is that some minor organic products may be lost or the yields of major products may be affected if they are unstable to these conditions. In this work, oxidative workups will be employed, unless otherwise specified, to screen for phenol and indene products and, to a lesser extent, for furans*I6and vinylcyclopentenedione' I products.

19

20

21

the alkyne. As indicated in Scheme 11, internal alkynes give mixtures of the two products, whereas terminal alkynes are chemoselective for the phenol products. This chemoselectivity of the reactions of the o-methoxy complex 2b with terminal alkynes has been taken advantage of in a number of synthetic applications,2,4A15 In planning syntheses such as those outlined in Scheme I, the preferred retroanalysis is back to a terminal acetylene since it has been firmly established that terminal acetylenes are incorporated with high regioselectivity" and since, as discussed above, key complexes such as the o-methoxy complex Zb are more chemoselective for phenol products with terminal alkynes than internal (15) (a) Semmelhack, M. F.; Bozell, J. J.; Sato, T.; Wulff, W. D.; Spiess, E.; Zask, A. J. Am. Chem. SOC.1982, 104, 5850. (b) Semmelhack, M.F.; Bozell, J. J.; Keller. L.; Sato, T.; Spiess, E. J.; Wulff, W. D.; Zask, A. Tetrahedron 1985, 41, 5803. (16) (a) Dotz, K. H. J. Organomet. Chem. 1977, 140, 177. (b) Wulff, W. D.; Gilbertson, S. R.; Springer, J. P. J . Am. Chem. SOC.1986, 108, 520. (c) McCallum, J . S.; Kunng, F. A.; Gilbertson, S. R.; Wulff, W. D. Orgonometallics 1988, 7, 2346. (17) (a) Wulff, W. D.; Tang, P. C.; McCallum, J. S.J. Am. Chem. Soc. 1981, 103, 7677. (b) Dbtz, K. H.; Muhlemeier, J.; Schubert, U.; Orama, 0. J. Organomet. Chem. 1983, 247, 187. (c) Yamashita, A,; Toy, A. Tetrohedron Left. 1986, 27, 3471

J. Am. Chem. SOC.,Vol. 113, No. 24, 1991 9295

Reactions of Aryl Chromium Carbene Complexes Scheme 111

0 Me0

1) nPrCiCH

(2 e q w THF, 45' C

+

2) Ce(Nhk(Nq)2

MeO ( c 9a 0 ) (0.5 5 c M) 3 i

r

MeO

OMe

0 49b

52b 16 YO

46 Yo

-wow 1

+

M&*m

OM8

nPr

nPr

L

J

53b 10 Yo

22

Scheme IV

2

2b 2b 2b 2a

E

OMe OMe OMe

H

LemP('C)

0.5

0.005 0.5 0.12

45 45 110 45

alkynes. However, it became readily apparent that a similar adaptation of the benzannulation of carbene complexes to the synthesis of olivin 7 (Scheme I) would not be as straightforward when it was found that the reaction of the 2,4-dimethoxyphenyl carbene complex 9a with I-pentyne was not chemoselective for the quinone 49b (Scheme 111, Table VIII). This reaction gave both the six- and five-membered ring annulation products 49b and 52b as well as keto ester 53b, which is a result of the oxidation of furan 22, a member of a known class of primary products of these reactions.8*'6 Clearly a reaction more chemoselective for phenol formation would be desirable in a planned synthesis of olivin. It was anticipated that the optimization of this reaction would be greatly aided by a deeper understanding of the reactions of arylcarbene complexes with acetylenes. Although the effects of the o-methoxy substituent in the reactions of complex 2b with alkynes were known, their origins had not been determined. Also complicating these reactions is the observation that we made 3 years ago that the product distribution from the reaction of the complex 2b with diethylacetylene is dependent on the concentration (Scheme IV).2f The yield of the quinone 14b falls from 61% to 5% when the concentration is dropped from 0.5 to 0.005 M. We further demonstrate here that the product distribution in the benzannulation reaction is dependent on the concentration of the alkyne and not on the concentration of the carbene complex.I8 We also report here the first observation of the dependence of the distribution between phenol and indene products on the reaction temperature. This effect is illustrated in Scheme IV in the reaction of 2b with diethylacetylene where the yield of quinone 14b drops from 61% to 29% when the temperature is raised from 45 to 110 OC. In the present work, the effect of concentration and temperature will both be carefully detailed for the reactions of carbene complexes and alkynes in general. In addition, the origin of the effect of the o-methoxy group on the partition between phenol and indene products will be examined. One of the overall goals of the work described herein is to further establish the synthetic scope of the reactions of arylcarbene (18) This has also been demonstrated for a molybdenum complex: Brandvold, T. A.; Chan, K. S.;Wulff, W. D.; Mitchell, J.; Clardy, J. C. To be submitted for publication.

14b 14b 14b 14a

61 % 5% 29% 00%

15b 15b 15b 1%

18% 66%

53% 5 0.3 YO

16b 5 % 16b 9 % 16b 1 0 % 16a S 0.3%

complexes with alkynes. In particular, this will be pursued with electron-rich arylcarbene complexes in an attempt to establish optimal procedures for application of the benzannulation reaction to the synthesis of olivin and 7-con-0-methylnogarol. The significance of this work will hopefully transcend our efforts toward the syntheses in Scheme I, since most polycyclic aromatic natural products contain predominantly electron-rich arene rings that are most typically substituted with oxygen. The other major goal of this work will be to gain a better understanding of the mechanism of the benzannulation reaction, with particular attention paid to the processes by which a variety of factors can influence the product distribution. Although the effect of alkyne concentration on product distribution has been previously observed in one particular reaction, the results to be found here more thoroughly define the scope of this phenomenon. The traditional mechanistic proposals for the benzannulation reaction cannot account for the effects of alkyne concentration nor can they account for many of the other factors observed in this study that influence product distribution, including temperature, solvent, and chelation effects. Specifically, it was the goal to identify, a t the very least, a mechanistic explanation that is consistent with all of the presently known facts concerning the benzannulation reaction for the purpose of guiding considerations of the implementation and strategic deployment of the benzannulation reaction in synthetic organic chemistry. Reactions of Monosubstituted Aryl Complexes with 3-Hexyne Before embarking directly on the optimization of the annulations of the dioxygenated complexes 5 and 9, it was decided to first identify the effects of a single oxygen substituent on the aryl ring. As mentioned above, the reactions of the o-methoxy complex 2b had been previously described, however, it had not been determined how the methoxyl group influences the product distribution.2e*2f~2J,4,s,'s The role of the methoxyl group on the product partition could be due to steric effects, to electron donation to the aryl ring by resonance, or by chelation of the oxygen of the methoxyl group to the metal at an intermediate along the pathway to either of the two products. T o distinguish between these possibilities, the series of monosubstituted complexes 2a-j were prepared and their reactions with 3-hexyne and 1-pentyne were examined (Scheme V).

Bos et ai.

9296 J . Am. Chem. SOC.,Vol. 113, No. 24, 1991

scheme VI Br

-

--

ii

i

WH3

R

(co’scr% 24

2C

RrCH20CH3

2d 20 21

RiCHflBu RmCHSH20CH3

\

82% 83 Yo

R=OtBu

25

““,,cj$

79% 79% CH30

/i OH

-

26

\ i

iv, v

_vi

OCH3

29

Br

73 Yo

84%

28

31

27

-

ef

viii

i, vii

43 70

83 Yo

oms 21

33

OAc

i, (a) nBuLi, (boCr(C0)6, (c) Me30BF4. ii, (a) tBuLi/cyclohexane, reflux, (b) Cr(C0)6, (c) Me30BF4. iii, heat or vacuum. iv, BH3. v, NaH, CHJ. vi, tBuOH, NaH, THF/DMF. vii, NaOCH,, CH30H. viii, Ac,O, pyridine, DMAP. Table I. Temperature and Solvent Effects in Reactions of 2b with 3-Hexyne‘ isolated yield (a) total mass 14b 1Sb 16b recovery (%) 0.5 61 18 5 84 53 10 92 0.5 29 0.5 13 38 18 69 2 5 84 PhH 0.5 11 6 31 12 0.5 29 12 16 12 44 0.5 heptane 0.5 81 14 1.2

The preparation of complexes 2b, 2i, and 2h were accomplished according to published procedure^.'^ The preparation of complex 2d has been improved by modification of the published procedure,“ which is indicated in Scheme VI. The complexes 2c, 2e, Zf,2g, and 2j are new, and all could be prepared in good yields as outlined in Scheme VI and as detailed in the Experimental Section. The only preparation that needs special mention is the p-acetoxy

complex 2j. Since p-bromophenyl acetate was not expected to undergo clean metal-halogen exchange with n-butyllithium, the p-acetoxy complex 2j was prepared in a two-step procedure from the silylated bromophenol 33, where the silyl group was replaced by acetate after the carbene complex was prepared.20 It has previously been reported that the o-methoxy complex 2b can be converted with slight heating to the tetracarbonyl chelated species

(19) Fischer, E.0.;Kreiter, C. G.; Kollmeier, H. J.; Muller, J.; Fischer, R. D.J . Organomet. Chem. 1971, 28, 231.

(20) For an alternate method of preparation of complexes of this typ, see: Semmelhack, M. F.; Lee, G. R. Organometallics 1987, 6, 1839.

Reactions of Aryl Chromium Carbene Complexes

J . Am. Chem. SOC.,Vol. 113, No. 24, 1991 9297

Table 11. Concentration and Solvent Effects in Reactions of 2b with 3-Hexyne isolated yield (%) solvent

THF

TH F

temp ("C)

45

1 IO

DMTHFh

45

CHJCN

45

PhH

45

PhH

1 IO

heptane

1 IO

[2](M) 0.5 0.05 0.005 0.05 0.005 0.05 0.5 0.5 0.05 0.005 0.05 0.005 0.5 0.005 0.5 0.005 0.5 0.05 0.005 0.5 0.005 0.5 0.05 0.005 0.5 0.5 0.5

[alkyne] 1.o 0.1

0.01 1 .o 1.o

8%

1 .o 1 .o 0.1 0.01 1 .o 1 .o 1.o 0.01 1 .o 0.01 1.o

0.1 0.01 1.o 0.01 1.o 0.1

0.01 1.o 1.o 1.o

time' 24 h 12 h 2 days 11 h 37 h 12 h 30 min 20 min

Ih 2h Ih 12 h 2h 2h 36 h 58 h 24 h 24 h 48 h 24 h 48 h 30 min 13 h 2h 50 min 25 min 25 min

14b 61 46 5

84 81

91 29 26 9 Trf 38

45 28 8 88 28

ISb

16b

18

5

42 66

1

5 5 4.0 45 52 c2 C 2 9 N N 61 a Et Et 0.5 N 0.14 N 88< 31 46 N 0.5 Tr 1 IO 11 N 88< 7 equiv) 0.5 M ceric ammonium nitrate solution in 0.1 N aqueous nitric acid, and the combined organic and aqueous layers were stirred for 20-30 min at room temperature (or where specified at 0 "C). After separation of the layers and extraction of the aqueous phase with additional ether, the combined organic extracts were washed with brine, dried (MgSO,), and concentrated. The 'H NMR spectrum of the crude reaction mixture was obtained to facilitate identification and isolation of components. (Typically, if the crude 'H NMR spectrum showed the presence of minor products, all solvent eluting from the column before or after the main bands came off was collected, concentrated, weighed, and examined by NMR in order to obtain a maximum limit for yield for these components.) Unless otherwise specified, the products of each reaction were separated by flash chromatography on silica gel with the eluent solvent mixture that is specified for each reaction. It is important to note that from the point of view of quinone synthesis it is not necessary to take the precaution of rigorously deoxygenating the reaction mixture by the freeze-thaw method. This was done to ensure that the product distributions were reproducible even for the very minor products. All of the reactions in all of the tables were carried out by the above general procedure unless specified in the table. For preparative purposes, these reactions give quinone products in many cases in yields that are the same whether or not the system is carefully deoxygenated. This is demonstrated in entries 7 and 8 in Table V and in entry 7 in Table VIII. For these reactions, the carbene complex was introduced into the reaction flask, which was then evacuated and filled with argon. The solvent was then added, which in the case of T H F was from a sodium benzophenone still and in the case of heptane was directly from a bottle as supplied by the Aldrich Chemical Co. The alkyne was added, the

threaded stopcock sealed, and the flask heated to the proper temperature. As can be seen from the data in Table V, the yield of quinone l l b did not seem to be affected by the level of oxygen introduced by this procedure. Reaction of o -Methoxyphenylcarbene Complex 2b with 3-Hexyne. The products obtained from these reactions are the indene 15b, the indenone 16b, the quinone 14b, and the cyclobutenone 72b, which were found to have spectral data identical with that reported previously for these compounds.2f The reactions under carbon monoxide atmosphere were performed by filling the evacuated reaction vessel in the last freeze-thaw cycle with carbon monoxide. A slightly positive pressure of CO was introduced at room temperature, and then the threaded stopcock was closed and the flask heated to 100 "C for the time indicated in Table 11. Only for the reaction of complex 2b with 3-hexyne was the quinone product observed to be sensitive to oxidation. In some reactions (180 OC and 0.5 M) the quinone 14b was obtained contaminated with an unknown oxidation product, 14b'. This product also formed upon chromatography on silica gel if elution was slow (14b and 14b' coelute and cannot be separated). Oxidation of 14b also occurs slowly standing in CDCI,. Spectral data for 14b': 'H NMR (CDCI,) 6 1.10 (t, 3 H, J = 7.5 Hz), 2.23-2.29 (m, 1 H), 2.37 (q, 2 H, J = 7.5 Hz), 2.57-2.61 (m, 1 H), 2.93-2.98 (m, 1 H), 2.98 (br s, 1 H, D 2 0exchangeable), 3.58-3.62 (m, 1 H), 3.91 (s, 3 H), 7.01 (d, 1 H, J = 8.1 Hz), 7.34 (t, 1 H, J = 8.0 Hz), 7.67 (d, 1 H, J = 7.8 Hz); IR (neat) 3457 sharp m, 2917 s, 2849 s, 1639 m, 1580 m, 1463 m, 1280 m, 1261 m cm-I; mass spectrum, m / e (re1 intensity) 260 M+ (IO), 244 (loo), 226 (59), 215 (65). 203 (92). The reaction in acetonitrile also produced the cyclobutenone 77b in amounts that varied with the conditions as indicated in Table 11. Cyclobutenone 77b had spectral data identical with those previously reported for this compound.2f This reaction was also carried out in the presence of varying amounts of isoprene and cyclopentadiene to examine the effects of these reagents on the distribution between phenol and indene products.* As can be seen from the data in Table XI, these reagents did not appear to have any influence on the distribution. Reaction of Complex 2b with 6-Dode~yne.2~ This reaction was carried out in THF as described in the general procedure, with 2 equiv of alkyne and a concentration of 2b of 0.005 M, and was complete within 20 min. After oxidative workup and chromatography with a 1:1:20 mixture of ether/CH2CI2/hexanesas eluent, this reaction gave the indene 36b (86%) as a colorless oil and indenone 37b (10%)as a yellow oil. Spectral data for 36b: Rf=0.26 (1:l:lO); 'H NMR (CDCI,) 6 0.90 (t, 6 H, J = 6.8 Hz), 1.2-1.6 (m, 12 H), 2.21-2.27 (m, 1 H), 2.36-2.40 (m, 1 H), 2.44 (t, 2 H, J = 7.7 Hz), 2.99 (s, 3 H), 3.90 (s, 3 H), 5.10 (s, 1 H), 6.70 (d, 1 H, J = 8.3 Hz), 6.81 (d, 1 H, J = 7.4 Hz), 7.24 (t, 1 H, J = 7.8 Hz); "C NMR (CDCI,) 6 (missing three aliphatic peaks due to overlap) 14.5, 23.0, 25.8, 26.1, 28.9, 29.6, 32.4, 52.0, 55.9, 81.7, 108.7, 112.3, 127.3, 130.5, 139.3, 143.4, 147.8, 156.5; IR (neat) 2955 s, 2929 s, 2858 m, 1605 m, 1588 m, 1478 m, 1260 s, 1084 m cm-I; mass spectrum, m/e (re1 intensity) 316 M+ (62), 259 (81), 245 (IOO), 189 (36), 171 (24). Anal. Calcd for C2,Hj202:C, 79.69; H, 10.19. Found: C, 79.45; H, 10.19. Spectral data for 37b: Rf= 0.14 (I:l:20); 'H NMR (CDCI,) 6 0.88 (t, 3 H, J = 6.8 Hz), 0.92 (t, 3 H, J = 6.8 Hz), 1.26-1.61 (m, 12 H), 2.22 (t, 2 H, J = 7.6 Hz), 2.48 (t, 2 H, J = 7.8 Hz), 3.92 (s, 3 H), 6.66 (d, 1 H, J = 7.1 Hz), 6.74 (d, 1 H, J = 8.6 Hz), 7.26 (dd, 1 H, J = 7.4 Hz, J = 8.4 Hz); "C NMR (CDCI,) 6 (missing two aliphatic peaks due to overlap) 14.4,22.9,23.2,26.6,28.5,29.4,32.2, 32.5, 56.3, 112.8, 113.8, 116.2, 135.5, 135.6, 148.7, 155.3, 156.2, 197.1; IR (neat) 2956 s, 2931 s, 2859 m, 1701 s, 1597 s, 1475 s, 1278 m, 1149 m, 1066 m cm-I; mass spectrum, m / e (re1 intensity) 300 M+ (15), 258 (4), 243 (47), 229 (IO), 187 (35), 178 (34), 137 (30), 124 (100); exact mass calcd for CZ0Hz8O2 ( m / e ) 300.2089, found ( m / e ) 300.2099. Reactions of o-Methylphenyl Complex 2g with 3-Hexyne. At high concentrations (Table 111) this reaction gives only the quinone 14g, which

9314 J. Am. Chem. SOC.,Vol. 113, No. 24, 1991 is obtained as a yellow crystalline solid (mp 81.5 "C). Spectral data for 14g: 'HNMR (CDCI,) 6 1.15 (t, 3 H, J = 7.5 Hz, CH2CHj), 1.16 (t, 3 H, J 7.5 Hz, CHZCH,), 2.63 (q, 2 H, J 7.5 Hz, CHZCH,), 2.64 (9. 2 H, J 7.4 Hz, CHZCH,), 2.75 (s, 3 H, CHj), 7.45 (d, I H, J = 7.4 Hz, H-6), 7.51 (t, 1 H, J = 7.6 Hz, H-7), 7.99 (d. 1 H, J = 7.7 Hz, H-8);"C NMR (CDCI,) 6 13.77, 13.94, 19.85, 20.20, 22.75, 124.86, 129.81, 132.91, 133.52, 137.08. 140.50, 146.19, 149.12, 185.16, 186.81; IR (CHCI,) 1653 s, 1590 w, 1326 m, 1279 m cm-I; mass spectrum, m/e (re1 intensity) 228 Mt (loo), 213 (36), 199 (IO), 185 (48), 171 (15), 141 (12), 128 (15). Anal. Calcd for CIJHI6O2:C, 78.91; H, 7.06. Found: C, 78.61; H, 7.03. At lower concentrations this reaction similarly gives the quinone 14g, but it also gives the indene 15g as a minor product that can be separated by flash chromatography on silica gel with a 1:1:30 mixture of ether/ CH2Clz/hexanesas eluent. The indene 15g was obtained as a yellow oil, and the following spectral data were collected: IH NMR (CDCI,) 6 1.1 1 (t, 3 H, J 7.6 Hz, CHZCH,), 1.12 (t, 3 H, J = 7.6 Hz, CHZCH,), 2.19-2.28 (m, 1 H, CHHCH,), 2.37 (s, 3 H, CH,), 2.40-2.46 (m, 3 H, CHHCH,, CHzCH,), 2.82 (s, 3 H, OCH,), 5.07 (s, 1 H, CHOCHj), 6.89 (d, I H, J = 7.6 Hz), 6.98 (d, 1 H, J = 7.3 Hz), 7.16 (t, 1 H, J = 7.5 Hz); IR (neat) 2966 vs, 2931 s, 2871 m, 1460 m, 1084 s cm-I; mass spectrum, m/e (re1 intensity) 216 M+ (65), 201 (18), 187 (IOO), 172 (27), 155 (19), 141 (16), 128 (17), 115 (13). Also isolated in a trace amount from these reactions was the indenone 16g (yellow oil), which precedes the quinone on TLC. Spectral data for 16g: 'H NMR (CDCI,) b 1.07 (t, 3 H, J = 7.5 Hz, CHZCH,), 1.22 (t, 3 H, J 7.6 Hz, CH2CHj), 2.27 (q, 2 H, J = 7.5 Hz, CHZCH,), 2.51 (s, 3 H, CH,), 2.54 (q, 2 H, J = 7.6 Hz, CH,CH,), 6.86 (d, 1 H, J = 7.2 Hz), 6.90 (d, 1 H, J = 8.2 Hz), 7.15 (t, 1 H, J = 7.4 Hz); IR (neat) 2969 s, 2933 s, 1700 vs, I594 s, I467 s, 1378 m, I259 m, 1 I57 m, 912 m, 785 m, 731 m cm-l. The reaction at 0.005 M and 1 IO OC produced both quinone and indene as indicated in Table 111, and while the isolated ratio of quinone to indene was 3.6:l, by crude 'H NMR spectra the ratio was approximately 2:l. Reactions of o-(Methoxymethy1)phenyl Complex 2c with 3-Hexyne. The reaction at 47 OC and 0.5 M was complete in 38 h, and separation of the products on silica gel with a 1:l:lO mixture of ether/CH2C12/ hexanes gave a fast-moving yellow compound, which by IH NMR was revealed to have not incorporated the carbene ligand and was probably an alkyne oligomer, but it was not further characterized. Further elution gave the quinone 14c in 74% yield as a yellow solid (mp 94-5 "C). Spectral data for 14c: RI= 0.27 (l:l:20); IH NMR (CDCIj) 6 1.14 (t, 6 H, J = 7.5 Hz, CHZCH,), 2.61 (q, 4 H, J = 7.3 Hz, CHZCH,), 3.53 (s, 3 H, CHZOCHj), 4.96 (s, 2 H, CH20CHj), 7.66 (t, 1 H, J 7.7 Hz, H-7), 7.99 (d, 1 H, J = 7.9 Hz, H-6), 8.01 (d, 1 H, J = 7.7 Hz, H-8); ',C NMR (CDCI,) 6 13.86, 13.99, 19.99, 20.25, 58.90, 72.91, 125.48, 128.52, 131.46, 132.92, 133.36, 141.82, 146.69, 148.88, 185.19, 187.08; 1R (CHCI,) 1653 vs, 1284 m, 1107 m cm-I; mass spectrum, m / e (re1 intensity) 258 M+ (17), 243 (27), 230 (49), 215 (100). Anal. Calcd for Cl6HI80,: C. 74.40 H, 7.02. Found: C, 74.13; H, 7.12. The reaction at 1 IO OC and 0.5 M was complete in 13.5 h and gave several products, which were separated on silica gel with a 1:l:lO mixture of ether/CH2CI2/hexanes. The products were identified as the indene 15g, which resulted from reduction of the o-methoxymethyl group and was obtained from the column contaminated with an alkyne oligomer (9% maximum yield of 15g), the quinone 14c in 59% yield, and finally the (methoxymethy1)indene1sC (Rj= 0.31, l:l:lO), which was obtained in 8% yield as a colorless oil. Spectral data for 1Sr: 'H NMR (CDCI,) 6 1.15 (t, 3 H, J = 7.5 Hz, CH$Hj), 1.16 (t, 3 H, J = 7.4 Hz, CH,CHj), 2.21-2.29 (m, 1 H, CHHCH3), 2.46-2.52 (m, 3 H, CH2CH3, CHHCH,), 2.86 (s, 3 H, OCHI), 3.45 (s, 3 H, OCH,), 4.56 (d, 1 H, J 12.1 Hz, CHHOCH,), 4.72 (d, 1 H, J 12.2 Hz, CHHOCH,), 5.15 s, I H, CHOCH&, 7.09 (d, 1 H, J 7.4 Hz), 7.18 (d, 1 H, J = 7.6 Hz), 7.28 (t, 1 H, J = 7.5 Hz); I3C NMR (CDC13) 6 13.60, 14.27, 18.41, 18.80, 50.32, 58.58,71.08, 81.48, 117.70, 123.91, 128.67, 134.83, 138.30, 140.80, 142.68, 144.89; IR (CHCI,) 2967 vs, 2932 vs, 2875 m, 1456 m, 1197 m, I102 vs, 1083 s cm-I; mass spectrum, m/e (re1 intensity) 246 Mt (60), 231 (9), 217 (loo), 199 ( I I ) , 185 (23), 171 (IO), 157 (19); exact mass calcd for CI6Hz2O2 (m/e) 246.1620, found (m/e) 246.1622. In the same manner, the reaction at 110 OC and 0.005 M in 14 h gave the reduced indene 15g (3.2-4.7%), an inseparable mixture containing the quinone 14c (6% yield), the indenone 16c (16% yield), and the enol ether 20c (9% yield), and finally, a fraction containing the indene 1% (24% maximum yield) that was contaminated with an unknown compound. Spectral data for 16c (yellow oil): IH NMR (CDCIj) 6 1.02 (t, 3 H, J 7.6 Hz, CHzCHj), 1.17 (t, 3 H, J = 7.6 Hz, CHZCHj), 2.22 (q, 2 H, J 7.6 Hz, CHzCHj), 2.59 (q, 2 H, J = 7.6 Hz, CHZCHj), 3.41 (S, 3 H, CHZOCH,), 4.79 (s, 2 H, CHZOCH,), 6.92 (dd, 1 H, J = 1.3 Hz, 3.4 Hz), 7.26 (d, 2 H, J = 3.3 Hz). Spectral data for 20c: partial 'H NMR data (derived from mixture, missing data for vinylic ethyl group due to overlap with resonances from indenone 16c) 'H NMR

Bos et al. (CDCI,) 6 0.54 (t, 3 H, CHHCH,), 1.76-1.87 (m, 1 H, CHHCH,), 1.98-2.08 (m, 1 H, CHHCH,), 3.32 (t, 1 H, CHCH2CHj). 3.43 (s, 3 H, OCHj), 3.80 (s, 3 H, OCHJ, 4.71 (s, 2 H, CHZOCH,), 7.11 (t, 1 H), 7.24-7.26 (m, 2 H). Reactions of 2-tert-Butoxy Complex 2d with 1Hexyne. This reaction was run under the three conditions indicated in Table IV and gave a distribution of the products 1 4 , E d , and 16d in the ratios that are indicated in Table IV. The quinone 14d could be separated by chromatography on silica gel with a 1:1:20 mixture of ether/CH2C12/hexanes as eluent, but the indene 1 9 and the indenone 16d could not be separated under these conditions and thus the yields in Table IV are determined on this mixture. The spectral data of 14d and 15d were found to be the same as those reported for these products from this reaction." The indenone 16d was not previously reported from this reaction but was observed in the present reactions at higher temperatures and was isolated as a yellow oil. Spectral data for 16d: 'H NMR (CDCI,) 6 1.06 (t. 3 H, J = 7.6 Hz), 1.22 (t, 3 H, J = 7.6 Hz), 1.44 (s, 9 H), 2.25 (q, 2 H, J = 7.5 Hz), 2.52 (4, 2 H, J = 7.6 Hz), 6.73 (d, 1 H, J = 7.1 Hz), 6.79 (d, 1 H, J = 8.5 Hz), 7.17 (t, 1 H, J = 7.6 Hz); IR (neat) 2973 s, 2934 m, 1703 s, 1592 s, 1462 s, 1 165 s, 867 m cm-I; mass spectrum, m/e (re1 intensity) 258 M+ ( 6 ) , 243 (12), 202 (IOO), 187 (95), 173 (IOO), 159 (21), 145 (22), 141 (17), 128 (20), 115 (45). Reactions of o-tert-Butoxymethyl Complex 2e with 3-Hexyne. At high concentration (0.5 M) this reaction produces the quinone 14e as the only detectable product in the yields indicated in Table IV. The quinone 14e was isolated as a yellow solid (mp 74-5 "C) and the following spectral data were obtained: 'H NMR (CDCI,) 6 1.12 (t, 3 H, J = 7.5 Hz. CHZCHj), 1.13 (t, 3 H, J = 7.5 Hz, CHZCH,), 1.31 (s, 9 H, C(CH,)j), 2.64 (q, 4 H, J = 7.5 Hz, CHZCH,), 4.96 (s, 2 H, CHzOtBu), 7.65 (t, 1 H, J = 7.8 Hz, H-7), 8.02 (d, 1 H, J = 7.6 Hz, H-6), 8.16 (d, 1 H, J = 7.9 Hz, H-8); "C NMR (CDCIj) 6 14.29, 14.46, 20.41, 20.66, 28.10, 62.78, 74.01, 125.53, 128.70, 132.32, 133.22, 133.55, 144.16, 146.94, 149.21, 185.69, 187.60; IR (CHCI,) 1651 vs, 1284 m, 1100 m cm-l; mass spectrum, m/e (re1 intensity) no parent was observed with either E1 or CI, showed loss of C4H, to give 244 (M+ - 56, IOO), 226 (26), 21 1 (17), 177 (21), 149 (25). Anal. Calcd for CI9H2,O3: C, 76.0; H, 8.10. Found: C, 75.85; H, 8.25. At low concentration (0.005 M) and 1IO OC, this reaction was complete in 1 h and produced several minor products. Chromatography on silica gel with a 1:1:20 mixture of ether/CH2Clz/hexanesgave an 8% yield of the indene 15g (from reduction of the terr-butoxy group) and an inseparable mixture of the quinone 14e (44% yield) and the indene 1% (1 3% yield). The ratio and yields of 14e and 15e were determined by integration of the benzylic protons. Also observed on occasion in varying but small amounts from this reaction was indenyl ether 2Oe. Spectral data for 15e (RI= 0.73 in 1:1:20, coelutes with quinone 14e): 'H NMR (CDCI,, derived from mixture) 6 -1.16 (m, 6 H), 1.32 (s, 9 H), 2.42-2.55 (m, 1 H), -2.64 (m, 3 H), 2.86 (s, 3 H), 4.56 (d, 1 H, J = 11.7 Hz), 4.67 (d, 1 H, J = 11.6 Hz), 4.82 (s, 1 H), 6.90 (d, 1 H, J = 7.0 Hz), 7.26 (t, 1 H, J = 7.5 Hz), 7.37 (d, 1 H, J = 8.2 Hz). Spectral data for 2Oe (light yellow oil, RI= 0.40, l:l:30): IH NMR (CDCI,) b 0.55 (t, 3 H, J = 7.4 Hz), 1.16 (t, 3 H, J = 7.5 Hz), 1.33 (s, 9 H), 1.76-1.81 (m, 1 H), 2.00-2.05 (m, 1 H), 2.14-2.18 (m, 1 H), 2.68-2.72 (m, 1 H), 3.33 (t, 1 H, J = 4.8 Hz), 3.81 (s, 3 H), 4.78 (s, 2 H), 7.13 (t, 1 H, J = 7.5 Hz), 7.20 (d, 1 H, J = 7.3 Hz), 7.35 (d, 1 H, J = 7.6 Hz); "C NMR (CDCI,) 6 8.28, 14.15, 18.03, 22.30, 27.75,45.66, 59.89, 60.93, 73.14, 121.57, 124.55, 126.63, 131.17, 133.53, 137.49, 145.09, 154.21; IR (neat) 2969 s, 2932 m, 2873 m, 1628 m, 1352 m, 1275 m, 1196 m, 1059 m, 766 m cm-l. Reaction of o-(Methoxyethyl) Complex 2f with 3-Hexyne. As indicated, this reaction at high concentrations gives the quinone 14f as the only isolable product. In the crude mixture from the reaction at 0.5 M and 1 IO "C, a trace amount of indene 15f was observed in the crude 'H NMR spectrum but could not be isolated. The reaction at 0.005 M and 1 IO OC gave three compounds that could not be separated upon chromatography on silica gel with a 1:1:20 mixture of ether/CH,Cl,/hexanes as eluent; they eluted as a single band. The yields in Table IV were determined by integration of the methoxyl groups of 14f, ISf, and 16f from this mixture. Also observed on occasion in varying but small amounts from this reaction was the indenyl ether 201. Analytical samples of 15f and 161 could be obtained by preparative TLC. Spectral data for 14f R / = 0.17 (l:l:20), yellow solid; mp 33-35 OC; IH NMR (CDCIj) 6 1.14 (t, 6 H, J = 7.5 Hz), 2.59-2.65 (m, 4 H), 3.35 (s, 3 H), 3.42 (t, 2 H, J = 6.4 Hz), 3.65 (t, 2 H, J = 6.5 Hz), 7.49-7.55 (m, 2 H), 8.00 (d, 1 H, J = 7.4 Hz); "C NMR (CDCI,) 6 13.78, 13.94, 19.87, 20.29, 35.23, 58.52, 72.46, 125.54, 129.93, 132.30, 133.86, 137.48, 141.17, 146.25, 149.35, 185.16, 186.92; IR (neat) 2972 m, 2935 m, 1 6 5 6 1623 ~ m, 1589 m, 1464 m, 1325 m, 1285 s, 1256 m, 11 14 s cm-I; mass spectrum, m / e (re1 intensity) 272 M+ (3), 240 (loo), 225 (74), 197 (17). Anal. Calcd for CI7HmO3:C, 74.97; H, 7.40. Found: C, 75.44; H, 7.59.

Reactions of Aryl Chromium Carbene Complexes

J . Am. Chem. SOC.,Vol. II3, No. 24, I991 9315

Spectral data for 15f R,= 0.17 (l:l:20),light yellow oil; 'H NMR H, J = 7.5 Hz, 8-H); "C NMR (CDCI,) 6 13.83,21.10,22.55,31.01, (CDCl,) 6 1.16 (s. 3 H, J = 7.2Hz), 1.17 (t, 3 H, J = 7.2 Hz), 2.24-2.29 125.42, 129.56,132.60,133.74, 136.49, 137.48, 140.85,149.66, 185.59, (m, 1 H), 2.46-2.51 (m, 3 H), 2.88 (s, 3 H), 2.96-3.00 (m, I H), 187.26;IR (CHCI,) 1656 vs cm-'; mass spectrum, m / e (re1 intensity) 3.15-3.19(m, I H), 3.38 (s, 3 H), 3.68(t, 2 H, J = 7.1 Hz). 5.14(s, 1 214 Mt (IOO), 199 (52),186 (34), 171 (55),158 (22), 143 (20), 128 H), 6.99 (d, 1 H, J = 7.6Hz), 7.03(d, 1 H, J = 7.4 Hz), 7.21 (t, 1 H, (49),115 (27). Anal. Calcd for C&& C, 78.48;H, 6.59. Found: J = 7.5 Hz); "C NMR (CDCI,) 8 13.62,14.28,18.38,18.80,31.84, C, 78.49;H, 6.70%. The crude 'H NMR from the reaction at 1 IO OC 50.24, 58.57, 72.56,81.62, 116.59, 125.92, 128.55, 135.61, 138.99, and 0.5 M showed one additional product besides quinone. Separation 140.78, 142.56,145.04;IR (neat) 2967 s, 2931 s, 2873 s, 2822 m, 1460 of the two products by chromatography on silica gel with a 1:l:lO mixture m, 11 16 s, 1083 s cm-'; decomposed upon attempted mass spectrum. of ether/CHzClz/hexanes gave a 77% yield of quinone Ilg (R,= 0.69) Spectral data for 16f:Rf= 0.17(l:l:20),yellow oil; 'H NMR (CDCI,) and a 7% yield of butenoate 40g (R, = 0.28),which was obtained as a 6 1.07(t, 3 H, J = 7.6Hz), 1.22 (t, 3 H, J = 7.6Hz), 2.26 (q, 2 H, J pale yellow oil. Spectral data for 40g: 'H NMR (CDCI,) 8 I .01 (t, 3 = 7.6 Hz), 2.53 (q, 2 H, J = 7.6 Hz), 3.21 (t, 2 H, J = 6.8Hz), 3.35 H, J 7.4 Hz, CHZCHZCH,), 1.59(q, 2 H, J = 7.5 Hz, CHZCHZCH,), (5. 3 H), 3.62 (t, 2 H, J = 6.7 Hz), 6.89 (d, 1 H, J = 7.1 Hz), 6.99(d, 2.41 (t, 2 H, J = 7.5 Hz, CHzCH2CH,), 2.54 (s, 3 H, CHI), 3.61 (s, 3 1 H, J = 7.8 Hz), 7.19 (t, 1 H, J = 7.5 Hz); IR (neat) 2968 m, 2928 H, COZCH,), 6.55(s, 1 H, C=CH), 7.22 (t, 1 H , J = 8.5 Hz), 7.35 (t, s, 2874 m, 1701 vs, 1592 m, 1466 m, 1 I 16 m cm-I; mass spectrum, m / e 1 H, J = 7.5 Hz), 7.56 (d, 1 H, J = 7.6 Hz); NMR (CDCI,) 8 13.6, (re1 intensity) 244 Mt (38),212 (loo), 197 (68),183 (99), 169 (19),155 20.7,20.9,36.1,52.0,125.5,129.5,130.5,131.6,131.7,137.1,138.7, (19), 141 (20), 128 (17), 115 (17). Spectral data for 2Of R,= 0.38 145.8,169.0,194.3;1R (neat) 2961 s, 2932 m, 1733 vs, 1700 m, 1671 (1:1:20),yellow oil; 'H NMR (CDCI,) 6 0.55 (t, 3 H, J = 7.4 Hz), 1.17 s, 1616 m, 1457 m, 1435 m, I258 s, 1210 s cm-I. (t, 3 H, J = 7.6 Hz), 1.76-1.82(m, 1 H), 2.00-2.05 (m, 1 H), 2.14-2.21 Reaction of the o-Methoxy Complex 2b with I-Pentyne. This reaction (m, 1 H), 2.68-2.76(m, 1 H), 3.12-3.25(m, 2 H), 3.33 (t, 1 H, J = 4.8 produces two products at high concentrations (0.5M, Table V) with THF Hz), 3.38 (s, 3 H), 3.61-3.63(m,2 H), 3.82 (s, 3 H), 7.04-7.08(m, 2 solvent as judged by the 'H NMR spectrum of the crude reaction mixH ), 7 . 1 8 (d, 1 H , J = 7 . 1 Hz);~~CNMR(CDCI~)88.22,14.18,18.14, tures, subsequent to the standard oxidative workup procedure. These two 22.34,32.07,45.58,58.53,60.85, 74.1I , 121.02,124.37,128.44,129.99, products were separated on a silica gel column with first a l:l:4mixture 132.67,138.47,145.53,154.23;mass spectrum, m / e (re1 intensity) 260 of ether/CH,Cl,/hexane as eluent and then with a l:l:2mixture to give M*(60), 231 (IOO), 228 (13),213 (35).199 (17),169 (IS); exact mass the keto ester 40b and the quinone l l b in the yields indicated in Table calcd for C17H2402( m / e ) 260.1776,found ( m / e ) 260.1780. V. The spectral data obtained for quinone l l b were identical with those Reactions of QMethoxy Complex 2h with 1Hexyne. The only product that have been reported for this compound.2r Spectral data for 40b that could be isolated and characterized from this reaction is the quinone (yellow oil): 'H NMR (CDCI,) 8 1.00 (t, 3 H, J = 7.4 1-k 14h for all of the reaction conditions indicated in Table IV. The spectral CH2CH2CH,), 1.58 (sextet, 2 H, J = 7.4 Hz, CH2CH2CH,), 2.39 (t, 2 data obtained for 14h were identical with those that have been previously H, J = 7.3 Hz, CHzCH2CH,), 3.73 (s, 3 H, OCH,), 3.88 (s, 3 H, reported for the product of this reaction." No five-membered ring OCH,), 6.74 (s, 1 H, C=CH), 6.93 (d, 1 H, J = 8.3 Hz), 6.99 (t, 1 H, products were observed in the crude IH NMR spectra for the reactions J = 7.6 Hz), 7.45 (dt, 1 H, J = 1.6 Hz, J = 8.5 Hz), 7.69 (dd, 1 H, J under any of the conditions indicated in Table IV, except for perhaps that = 1.6 Hz, J = 7.7 Hz); "C NMR (CDCI,) 8 13.43,20.60,36.17,45.62, at 45 OC and 0.5 M. In the workup of this reaction, a small amount of 52.04,111.63,120.75,127.80,130.42,130.99,133.88,144.67,158.75, yellow material followed the quinone off the column with a 1:1:20mix169.98, 191.05;IR (neat) 1732 vs, 1661 m, 1614 m, 1598 m, 1485 m, ture of ether/CH2CIz/hexane as solvent, and despite the fact that this 1465 m, 1437 m, 1293 s, 1254 s, 1242 s, 1203 s cm-I; mass spectrum, fraction was a mixture of many components, absorptions that would have m/e (re1 intensity) 262 Mt (13), 231 (18),201 (20), 187 (13), 173 (12), been anticipated for the indenone 16h were present, but based on the I55(8), 135 (IOO), 121 (8), llO(8),92(l5),77(32);exactmasscalcd weight of this fraction the maximum yield would be 5%. for C15H1804 ( m / e ) 262.1205,found ( m / e )262.1205. As indicated in Table V, the reactions at high concentrations with benzene as solvent Reactions of 3-Methoxy Complex 2i with 3-Hexyne. The benzannulation reaction with meta-substituted arylcarbene complexes can give produce the quinone I l b but not the keto ester 40b. In the benzene two possible regioisomers. This aspect of the regiochemistry of the reactions it is anticipated that cyclopentenedione products are likely reaction has been briefly examined previously, and it was found that the formed as minor products but are not detected since these products are selectivity was dependent on the electronic nature of the meta substitudestroyed by Ce'V.ll ent.". The reaction of the m-methoxyphenyl complex 2i with 3-hexyne The reaction at 1IO OC and 0.005 M was done in 35 min and gave has been previously reported"' to give a 2.1:l.O mixture of the quinones four products. These could be separated by chromatography on silica gel 14h and 14b, and we find here that this regioselectivity is only slightly with gradient elution, beginning with a 1:l:lO mixture of ether/ affected by temperature or concentration, as indicated by the data in CH,Cl,/hexanes, then with a l:l:8 mixture, and finally with a l:l:4 Table IV. These reactions were unusually clean and no evidence for mixture to give a 13% yield of the indene 12b, a 6% yield of slightly five-membered rings or other minor products could be gleaned from the impure keto ester 40b, a 23% yield of the quinone Ilb, and a 28% yield crude 'H NMR spectra or the TLCs of these reactions. of the indanone 38b. Spectral data for indene 12b (colorless oil, R, = Reactiom of QAcetoxy Complex 2j with 1Hexyne. At 45 OC and 0.5 0.54 in l:l:4): IH NMR (CDCI,) 8 1.00 (t, 3 H, J = 7.3 Hz, M, this reaction produced a 94.4% yield of quinone 14j along with a small CH2CHzCH,), 1.59-1.69(m, 2 H, CH2CHzCH,), 2.34 (t, 2 H, J = 7.4 amount of another colored annulated product (presumably indenone), Hz, CH$H2CH,), 3.04 (s, 3 H, CHOCH,), 3.90(s, 3 H, OCH,), 5.06 which was observed but not isolated. Spectral data for 14j (yellow solid, (s, 1 H, CHOCH3),6.37 (s, 1 H, C(H)=CnPr), 6.68 (d, 1 H, J = 8.3 mp 68-70 "C): 'H NMR (CDCIJ 8 1.15 (t. 6 H, J = 7.5 Hz, CH,CH,), Hz), 6.78 (d, 1 H, J = 7.3 Hz), 7.20 (t, 1 H, J = 7.8 Hz); I3C NMR 2.34 (s, 3 H, COCH,), 2.636 (q, 2 H, J 7.5 Hz, CHZCH,), 2.640 (q, (CDCI,) 8 14.04, 21.40, 30.29, 51.80, 55.39, 82.43, 108.25, 113.47, 2 H, J = 7.5 Hz, CH$H,), 7.36 (dd, 1 H, J = 2.3,8.6 Hz, H-6). 7.74 126.84,126.87,130.27,145.85,150.79,156.17;IR (neat) 2959 m, 2930 (d, 1 H, J = 2.3 Hz, H-5), 8.08 (d, 1 H, J = 8.3 Hz, H-8); "C NMR m, 1480 m, 1261 m, 1089 m, 909 s, 734 s cm-I; mass spectrum, m/e (re1 (CDCIj) 6 13.90,13.93,20.10,20.15,21.04,119.20,126.52,128.22, intensity) 218 Mt (38),189 (IOO), 175 (33), 115 (23);exact mass calcd 129.81, 133.79, 148.22, 148.16, 154.55, 168.64, 184.06, 183.96;IR for CI4Hl8OZ ( m / e ) 218.1307,found ( m / e ) 218.1291. Spectral data for (CHCI,) 2877 m, 1760 vs, 1667 vs, 1614 s, 1594 s, 1462 m, 1369 m, 1345 indanone 38b (yellow oil, R, = 0.24 in l:l:4): 'H NMR (CDCI,) 6 0.95 m, 1320 m, 1136 m, 987 m, 948 m, 867 m cm-I; mass spectrum, m / e (re1 (t. 3 H, J = 7.1 Hz, CHzCH2CH3),1.36-1.48 (m, 3 H, CHHCH2CH3), intensity) 272 M+ (38),230 (loo), 215 (27),83 (64). Anal. Calcd for 1.83-1.88(m, 1 H, CHHCH,CH,), 2.34 (dd, 1 H, J = 3.3 Hz, J = 18.8 Cl6HI6o4:c. 70.58;H, 5.92. Found: C, 70.42;H, 6.00. The crude Hz, COCHHCH(nPr)), 2.81 (dd, 1 H, J = 7.7 Hz, J = 18.8 Hz, mixture from the reaction at 1 IO OC and 0.5M was subjected to oxiCOCHHCH(nPr)), 3.25-3.29 (m. 1 H, COCH,CH(nPr)), 3.93 (s, 3 H, dation and gradient elution chromatography, first with a l:l:8 mixture OCH,), 6.75 (d, 1 H, J 8.2Hz, C=CH), 7.01 (d, I H, J = 7.5 Hz), of ether/CH2C12/hexanesand then a l:l:4 mixture. After a colored band 7.50 (t, 1 H, J = 7.9 Hz); ',C NMR (CDCI,) 8 14.10,20.61, 37.58, containing alkyne oligomers eluted, a fraction containing an 80.4% yield 38.33,43.58,55.73, 108.87, 117.32, 124.79, 136.29, 157.82, 161.81, of quinone 14j was obtained. Three other minor products were collected 204.20;IR (neat) 2958 m, 2930 m, 1708 vs, 1593 s, 1480 s, 1280 m, 1229 each in an impure state, two of which are probably aldehydes (based on m, 1027 m cm-I; mass spectrum, m / e (re1 intensity) 204 Mt (IOO), 175 IH NMR shifts at 9.6and 9.7ppm); however, no compounds with spectra (61), 162 (65). 144 (20). 115 (19), 103 (23);exact mass calcd for C13expected for the indenyl or indenone products were observed. Hl6OZ( m / e ) 204.1150,found ( m / e ) 204.1142. Reaction of Complex 2g with 1-Pentyne. As indicated in Table V, this Reactions of o-(Methoxymethyl) Complex 2c with I-Pentyne. Under reaction at high concentrations (0.5M) gave exclusively the quinone l l g conditions of low temperature (45 "C) and high concentration (0.5M) as a yellow solid (mp 63-65 "C), for which the following spectral data this reaction goes to completion in 14 h and produces only the quinone were obtained: 'H NMR (CDCI,) 6 1.01 (t, 3 H, J = 7.3 Hz, Ilc in 56% yield, as indicated in Table V. Spectral data for quinone l l c CH,CH,CH,), 1.62 (q, 2 H, J = 7.4 Hz, CH2CH2CH,), 2.52 (t, 2 H, (yellow solid, mp 71-72 OC): 'H NMR (CDCI,) 8 1.02 (t, 3 H, J = 7.4 J 7.6Hz, CH2CH2CH3). 2.75 (s, 3 H, CH3), 6.69 (s, 1 H, 3-H), 7.48 Hz, CHZCHZCH,), 1.62(q, 2 H, J = 7.5 Hz, CH,CH,CH,), 2.53 (t, 2 (d, 1 H, J = 7.4 Hz, 6-H), 7.55 (t, 1 H, J = 7.6 Hz, 7-H), 8.00 (d, 1 H, J = 7.3Hz, CHzCH2CH3), 3.55 (s, 3 H, CH2OCH3), 4.99 (s, 2 H,

9316 J . Am. Chem. SOC.,Vol. 113. No. 24, 1991

Bos et al.

Table XII. Effect of Lewis Acids on the Reaction of Comdex 2b with I-Pentvne' isolated yield (%)* total mass [2h] (M) additive (equiv) Ilh 38b [alkyne] 40b recovery (W) six/five 1 IO 0.5 1.o none 41 39 13 93 1.1 0.5 1 .o tBuMezSiCI (1.5) 46 28 N 74 1.6 heptane 1 IO 0.5 1 .O none 50 23 2 75 2.0 0.5 1.o tBuMezSiC1 (1.5) 55 26 2 83 2.0 0.5 I .o Me,SiCI (1.5) 52 22 N 74c 2.4 0.5 1.o AcZO (1.5) 26 12 9 47d 1.2 0.5 1.o BFvOEt, (1.5) 56 20 N 76 2.8 OReaction time is 25 min in each case. b N indicates product not detected by 500 MHz 'H NMR of the crude reaction mixture.